训练问题 #12
Comments
|
你好，你是直接下载代码后运行的吗？没有进行更改吗？
banxianerr ***@***.***> 于2022年5月16日周一 20:24写道：
… 老师你好，我在训练时使用的是这个命令：python demo.py --dataset='Indian' --epoches=300
--patches=7 --band_patches=3 --mode='CAF'
--weight_decay=5e-3，但是跑出来的精度只有75%，方便帮我看一下可能是哪里出现的问题吗？
[image: image]
<https://user-images.githubusercontent.com/62459853/168591652-e606d2c6-17cf-41ec-beff-715eb35d7cc4.png>
—
Reply to this email directly, view it on GitHub
<#12>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AFL2GZT52DQE3GWOU374MADVKI47PANCNFSM5WBJ3ROQ>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
|
如果是直接从github下载直接run，我们刚测试了下是没有问题的
Danfeng Hong ***@***.***> 于2022年5月17日周二 17:46写道：
… 你好，你是直接下载代码后运行的吗？没有进行更改吗？
banxianerr ***@***.***> 于2022年5月16日周一 20:24写道：
> 老师你好，我在训练时使用的是这个命令：python demo.py --dataset='Indian' --epoches=300
> --patches=7 --band_patches=3 --mode='CAF'
> --weight_decay=5e-3，但是跑出来的精度只有75%，方便帮我看一下可能是哪里出现的问题吗？
> [image: image]
> <https://user-images.githubusercontent.com/62459853/168591652-e606d2c6-17cf-41ec-beff-715eb35d7cc4.png>
>
> —
> Reply to this email directly, view it on GitHub
> <#12>, or
> unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AFL2GZT52DQE3GWOU374MADVKI47PANCNFSM5WBJ3ROQ>
> .
> You are receiving this because you are subscribed to this thread.Message
> ID: ***@***.***>
>
|
老师，这是我的整个操作流程，是直接下载然后run的 patch is : 7 x_train shape = (695, 7, 7, 200), type = float64 x_train_band shape = (695, 147, 200), type = float64 y_train: shape = (695,) ,type = int64 start training Final result: Parameter: |
|
如果没有改变任何代码，有可能是因为python版本之类的问题（我们使用的版本也在github说明了），因为我们这个地方是把初始化的seed固定了，之前也有其他人问过我们这个问题，但他们的结果是比我们report还要高的。
banxianerr ***@***.***> 于2022年5月17日周二 20:21写道：
… 你好，你是直接下载代码后运行的吗？没有进行更改吗？ banxianerr *@*.
*> 于2022年5月16日周一 20:24写道： … <#m_-1848683633840172713_>
老师你好，我在训练时使用的是这个命令：python demo.py --dataset='Indian' --epoches=300
--patches=7 --band_patches=3 --mode='CAF'
--weight_decay=5e-3，但是跑出来的精度只有75%，方便帮我看一下可能是哪里出现的问题吗？ [image: image]
https://user-images.githubusercontent.com/62459853/168591652-e606d2c6-17cf-41ec-beff-715eb35d7cc4.png
<https://user-images.githubusercontent.com/62459853/168591652-e606d2c6-17cf-41ec-beff-715eb35d7cc4.png>
— Reply to this email directly, view it on GitHub <#12
<#12>>, or
unsubscribe
https://github.com/notifications/unsubscribe-auth/AFL2GZT52DQE3GWOU374MADVKI47PANCNFSM5WBJ3ROQ
<https://github.com/notifications/unsubscribe-auth/AFL2GZT52DQE3GWOU374MADVKI47PANCNFSM5WBJ3ROQ>
. You are receiving this because you are subscribed to this thread.Message
ID: @.*>
老师，这是我的整个操作流程，是直接下载然后run的
(gml-nts) ***@***.***:/gml-project/test$ git clone
https://github.com/danfenghong/IEEE_TGRS_SpectralFormer.git
正克隆到 'IEEE_TGRS_SpectralFormer'...
remote: Enumerating objects: 111, done.
remote: Counting objects: 100% (6/6), done.
remote: Compressing objects: 100% (6/6), done.
remote: Total 111 (delta 2), reused 2 (delta 0), pack-reused 105
接收对象中: 100% (111/111), 13.87 MiB | 54.00 KiB/s, 完成.
处理 delta 中: 100% (45/45), 完成.
(gml-nts) ***@***.***:/gml-project/test$ ls
IEEE_TGRS_SpectralFormer
(gml-nts) ***@***.***:/gml-project/test$ cd IEEE_TGRS_SpectralFormer/
(gml-nts) ***@***.***:/gml-project/test/IEEE_TGRS_SpectralFormer$ ls
data demo.py log README.md SpectralFormer.PNG vit_pytorch.py
(gml-nts) ***@***.***:~/gml-project/test/IEEE_TGRS_SpectralFormer$ python
demo.py --dataset='Indian' --epoches=300 --patches=7 --band_patches=3
--mode='CAF' --weight_decay=5e-3
height=145,width=145,band=200
------------------------------
patch is : 7
mirror_image shape : [151,151,200]
------------------------------
x_train shape = (695, 7, 7, 200), type = float64
x_test shape = (9671, 7, 7, 200), type = float64
x_true shape = (21025, 7, 7, 200), type = float64
------------------------------
x_train_band shape = (695, 147, 200), type = float64
x_test_band shape = (9671, 147, 200), type = float64
x_true_band shape = (21025, 147, 200), type = float64
------------------------------
y_train: shape = (695,) ,type = int64
y_test: shape = (9671,) ,type = int64
y_true: shape = (21025,) ,type = int64
------------------------------
start training
/home/GaoML/anaconda3/envs/gml-nts/lib/python3.6/site-packages/torch/optim/lr_scheduler.py:136:
UserWarning: Detected call of lr_scheduler.step() before optimizer.step().
In PyTorch 1.1.0 and later, you should call them in the opposite order:
optimizer.step() before lr_scheduler.step(). Failure to do this will
result in PyTorch skipping the first value of the learning rate schedule.
See more details at
https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate
"https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate",
UserWarning)
Epoch: 001 train_loss: 2.8026 train_acc: 6.7626
Epoch: 002 train_loss: 2.7212 train_acc: 8.2014
Epoch: 003 train_loss: 2.6631 train_acc: 14.6763
Epoch: 004 train_loss: 2.4813 train_acc: 19.4245
Epoch: 005 train_loss: 2.2231 train_acc: 23.3094
Epoch: 006 train_loss: 2.0518 train_acc: 29.9281
Epoch: 007 train_loss: 1.9562 train_acc: 32.0863
Epoch: 008 train_loss: 1.8885 train_acc: 32.8058
Epoch: 009 train_loss: 1.8270 train_acc: 35.8273
Epoch: 010 train_loss: 1.7620 train_acc: 38.2734
Epoch: 011 train_loss: 1.6822 train_acc: 39.4245
Epoch: 012 train_loss: 1.5872 train_acc: 44.4604
Epoch: 013 train_loss: 1.5285 train_acc: 43.7410
Epoch: 014 train_loss: 1.4653 train_acc: 47.1942
Epoch: 015 train_loss: 1.4087 train_acc: 48.2014
Epoch: 016 train_loss: 1.3532 train_acc: 50.0719
Epoch: 017 train_loss: 1.2949 train_acc: 52.2302
Epoch: 018 train_loss: 1.2636 train_acc: 54.3885
Epoch: 019 train_loss: 1.2534 train_acc: 53.2374
Epoch: 020 train_loss: 1.1967 train_acc: 54.6763
Epoch: 021 train_loss: 1.2165 train_acc: 52.6619
Epoch: 022 train_loss: 1.1935 train_acc: 55.5396
Epoch: 023 train_loss: 1.1406 train_acc: 57.5540
Epoch: 024 train_loss: 1.1320 train_acc: 56.5468
Epoch: 025 train_loss: 1.1173 train_acc: 57.4101
Epoch: 026 train_loss: 1.1210 train_acc: 56.5468
Epoch: 027 train_loss: 1.1173 train_acc: 56.8345
Epoch: 028 train_loss: 1.0976 train_acc: 57.8417
Epoch: 029 train_loss: 1.0393 train_acc: 59.4245
Epoch: 030 train_loss: 1.0321 train_acc: 59.8561
Epoch: 031 train_loss: 1.0060 train_acc: 62.1583
Epoch: 032 train_loss: 1.0072 train_acc: 60.8633
Epoch: 033 train_loss: 1.0057 train_acc: 61.2950
Epoch: 034 train_loss: 0.9799 train_acc: 63.0216
Epoch: 035 train_loss: 0.9426 train_acc: 67.0504
Epoch: 036 train_loss: 0.9399 train_acc: 64.1727
Epoch: 037 train_loss: 0.9494 train_acc: 63.8849
Epoch: 038 train_loss: 0.9657 train_acc: 62.4460
Epoch: 039 train_loss: 0.9423 train_acc: 65.4676
Epoch: 040 train_loss: 0.9430 train_acc: 64.7482
Epoch: 041 train_loss: 0.9671 train_acc: 61.1511
Epoch: 042 train_loss: 0.9029 train_acc: 63.7410
Epoch: 043 train_loss: 0.8931 train_acc: 65.1799
Epoch: 044 train_loss: 0.9375 train_acc: 62.5899
Epoch: 045 train_loss: 0.9465 train_acc: 62.7338
Epoch: 046 train_loss: 0.9501 train_acc: 61.8705
Epoch: 047 train_loss: 0.9081 train_acc: 65.0360
Epoch: 048 train_loss: 0.8632 train_acc: 66.1870
Epoch: 049 train_loss: 0.9113 train_acc: 63.4532
Epoch: 050 train_loss: 0.8629 train_acc: 66.9065
Epoch: 051 train_loss: 0.8375 train_acc: 67.1942
Epoch: 052 train_loss: 0.8520 train_acc: 69.6403
Epoch: 053 train_loss: 0.8167 train_acc: 68.2014
Epoch: 054 train_loss: 0.8239 train_acc: 68.0576
Epoch: 055 train_loss: 0.8760 train_acc: 66.3309
Epoch: 056 train_loss: 0.8432 train_acc: 67.7698
Epoch: 057 train_loss: 0.8148 train_acc: 69.0648
Epoch: 058 train_loss: 0.7857 train_acc: 71.5108
Epoch: 059 train_loss: 0.8108 train_acc: 68.3453
Epoch: 060 train_loss: 0.7942 train_acc: 69.6403
Epoch: 061 train_loss: 0.7868 train_acc: 70.2158
Epoch: 062 train_loss: 0.7558 train_acc: 72.9496
Epoch: 063 train_loss: 0.7878 train_acc: 71.3669
Epoch: 064 train_loss: 0.7764 train_acc: 71.3669
Epoch: 065 train_loss: 0.7436 train_acc: 71.9424
Epoch: 066 train_loss: 0.7538 train_acc: 69.6403
Epoch: 067 train_loss: 0.7628 train_acc: 70.6475
Epoch: 068 train_loss: 0.7292 train_acc: 70.5036
Epoch: 069 train_loss: 0.7443 train_acc: 72.6619
Epoch: 070 train_loss: 0.7410 train_acc: 73.6691
Epoch: 071 train_loss: 0.7131 train_acc: 74.2446
Epoch: 072 train_loss: 0.6870 train_acc: 73.6691
Epoch: 073 train_loss: 0.7287 train_acc: 71.6547
Epoch: 074 train_loss: 0.8080 train_acc: 66.7626
Epoch: 075 train_loss: 0.7769 train_acc: 70.0719
Epoch: 076 train_loss: 0.6967 train_acc: 73.5252
Epoch: 077 train_loss: 0.6612 train_acc: 75.9712
Epoch: 078 train_loss: 0.6892 train_acc: 75.5396
Epoch: 079 train_loss: 0.6764 train_acc: 74.6763
Epoch: 080 train_loss: 0.6704 train_acc: 75.9712
Epoch: 081 train_loss: 0.6446 train_acc: 76.2590
Epoch: 082 train_loss: 0.6633 train_acc: 74.8201
Epoch: 083 train_loss: 0.6934 train_acc: 72.8058
Epoch: 084 train_loss: 0.6892 train_acc: 71.9424
Epoch: 085 train_loss: 0.6701 train_acc: 73.3813
Epoch: 086 train_loss: 0.6569 train_acc: 74.9640
Epoch: 087 train_loss: 0.6166 train_acc: 76.8345
Epoch: 088 train_loss: 0.6207 train_acc: 76.5468
Epoch: 089 train_loss: 0.6254 train_acc: 78.1295
Epoch: 090 train_loss: 0.6029 train_acc: 77.4101
Epoch: 091 train_loss: 0.6522 train_acc: 75.1079
Epoch: 092 train_loss: 0.6247 train_acc: 75.6834
Epoch: 093 train_loss: 0.5812 train_acc: 79.2806
Epoch: 094 train_loss: 0.5894 train_acc: 77.1223
Epoch: 095 train_loss: 0.6022 train_acc: 77.9856
Epoch: 096 train_loss: 0.5708 train_acc: 80.0000
Epoch: 097 train_loss: 0.5900 train_acc: 78.7050
Epoch: 098 train_loss: 0.5914 train_acc: 77.1223
Epoch: 099 train_loss: 0.5767 train_acc: 78.7050
Epoch: 100 train_loss: 0.5974 train_acc: 77.5540
Epoch: 101 train_loss: 0.5525 train_acc: 80.1439
Epoch: 102 train_loss: 0.5511 train_acc: 79.8561
Epoch: 103 train_loss: 0.5534 train_acc: 81.1511
Epoch: 104 train_loss: 0.5484 train_acc: 79.8561
Epoch: 105 train_loss: 0.5108 train_acc: 82.8777
Epoch: 106 train_loss: 0.5144 train_acc: 81.0072
Epoch: 107 train_loss: 0.5461 train_acc: 80.5755
Epoch: 108 train_loss: 0.5080 train_acc: 83.7410
Epoch: 109 train_loss: 0.5013 train_acc: 81.1511
Epoch: 110 train_loss: 0.5130 train_acc: 80.8633
Epoch: 111 train_loss: 0.5016 train_acc: 80.8633
Epoch: 112 train_loss: 0.4976 train_acc: 84.0288
Epoch: 113 train_loss: 0.4889 train_acc: 83.4532
Epoch: 114 train_loss: 0.5491 train_acc: 78.4173
Epoch: 115 train_loss: 0.5535 train_acc: 78.9928
Epoch: 116 train_loss: 0.4967 train_acc: 80.2878
Epoch: 117 train_loss: 0.4734 train_acc: 84.4604
Epoch: 118 train_loss: 0.4810 train_acc: 83.7410
Epoch: 119 train_loss: 0.4544 train_acc: 85.7554
Epoch: 120 train_loss: 0.4862 train_acc: 84.1727
Epoch: 121 train_loss: 0.4977 train_acc: 80.8633
Epoch: 122 train_loss: 0.4506 train_acc: 83.7410
Epoch: 123 train_loss: 0.4349 train_acc: 84.4604
Epoch: 124 train_loss: 0.4409 train_acc: 85.6115
Epoch: 125 train_loss: 0.4298 train_acc: 87.0504
Epoch: 126 train_loss: 0.4384 train_acc: 84.4604
Epoch: 127 train_loss: 0.4050 train_acc: 86.9065
Epoch: 128 train_loss: 0.4319 train_acc: 85.4676
Epoch: 129 train_loss: 0.4201 train_acc: 86.6187
Epoch: 130 train_loss: 0.3926 train_acc: 85.4676
Epoch: 131 train_loss: 0.4054 train_acc: 84.7482
Epoch: 132 train_loss: 0.4109 train_acc: 84.0288
Epoch: 133 train_loss: 0.3931 train_acc: 87.1942
Epoch: 134 train_loss: 0.4001 train_acc: 87.0504
Epoch: 135 train_loss: 0.4284 train_acc: 85.8993
Epoch: 136 train_loss: 0.4334 train_acc: 85.3237
Epoch: 137 train_loss: 0.3835 train_acc: 86.1870
Epoch: 138 train_loss: 0.4041 train_acc: 85.6115
Epoch: 139 train_loss: 0.3936 train_acc: 87.1942
Epoch: 140 train_loss: 0.4148 train_acc: 86.7626
Epoch: 141 train_loss: 0.3784 train_acc: 86.9065
Epoch: 142 train_loss: 0.3477 train_acc: 88.2014
Epoch: 143 train_loss: 0.3728 train_acc: 87.9137
Epoch: 144 train_loss: 0.3487 train_acc: 90.0719
Epoch: 145 train_loss: 0.3473 train_acc: 88.7770
Epoch: 146 train_loss: 0.3521 train_acc: 88.6331
Epoch: 147 train_loss: 0.3403 train_acc: 89.2086
Epoch: 148 train_loss: 0.4209 train_acc: 84.8921
Epoch: 149 train_loss: 0.3766 train_acc: 88.0576
Epoch: 150 train_loss: 0.4179 train_acc: 85.7554
Epoch: 151 train_loss: 0.3921 train_acc: 85.7554
Epoch: 152 train_loss: 0.3338 train_acc: 90.3597
Epoch: 153 train_loss: 0.3174 train_acc: 90.6475
Epoch: 154 train_loss: 0.3225 train_acc: 91.5108
Epoch: 155 train_loss: 0.3466 train_acc: 88.3453
Epoch: 156 train_loss: 0.3500 train_acc: 89.2086
Epoch: 157 train_loss: 0.3596 train_acc: 88.0576
Epoch: 158 train_loss: 0.3281 train_acc: 89.9281
Epoch: 159 train_loss: 0.3289 train_acc: 90.0719
Epoch: 160 train_loss: 0.3189 train_acc: 89.4964
Epoch: 161 train_loss: 0.3102 train_acc: 90.9352
Epoch: 162 train_loss: 0.2979 train_acc: 91.5108
Epoch: 163 train_loss: 0.2798 train_acc: 91.5108
Epoch: 164 train_loss: 0.2891 train_acc: 91.0791
Epoch: 165 train_loss: 0.3358 train_acc: 88.6331
Epoch: 166 train_loss: 0.3216 train_acc: 90.2158
Epoch: 167 train_loss: 0.3456 train_acc: 89.7842
Epoch: 168 train_loss: 0.3664 train_acc: 87.0504
Epoch: 169 train_loss: 0.3508 train_acc: 87.6259
Epoch: 170 train_loss: 0.3234 train_acc: 89.4964
Epoch: 171 train_loss: 0.2797 train_acc: 91.5108
Epoch: 172 train_loss: 0.2651 train_acc: 93.0935
Epoch: 173 train_loss: 0.2886 train_acc: 90.5036
Epoch: 174 train_loss: 0.2518 train_acc: 92.0863
Epoch: 175 train_loss: 0.2634 train_acc: 91.2230
Epoch: 176 train_loss: 0.2718 train_acc: 91.3669
Epoch: 177 train_loss: 0.2945 train_acc: 89.9281
Epoch: 178 train_loss: 0.2711 train_acc: 92.2302
Epoch: 179 train_loss: 0.2721 train_acc: 91.7986
Epoch: 180 train_loss: 0.2767 train_acc: 91.7986
Epoch: 181 train_loss: 0.2502 train_acc: 92.8058
Epoch: 182 train_loss: 0.2626 train_acc: 91.2230
Epoch: 183 train_loss: 0.2774 train_acc: 90.6475
Epoch: 184 train_loss: 0.2583 train_acc: 92.5180
Epoch: 185 train_loss: 0.2687 train_acc: 92.2302
Epoch: 186 train_loss: 0.2641 train_acc: 91.2230
Epoch: 187 train_loss: 0.2673 train_acc: 92.0863
Epoch: 188 train_loss: 0.3005 train_acc: 88.7770
Epoch: 189 train_loss: 0.2642 train_acc: 91.6547
Epoch: 190 train_loss: 0.2427 train_acc: 93.5252
Epoch: 191 train_loss: 0.1970 train_acc: 95.6834
Epoch: 192 train_loss: 0.2024 train_acc: 94.5324
Epoch: 193 train_loss: 0.2051 train_acc: 94.2446
Epoch: 194 train_loss: 0.2177 train_acc: 93.9568
Epoch: 195 train_loss: 0.2230 train_acc: 93.3813
Epoch: 196 train_loss: 0.2175 train_acc: 93.2374
Epoch: 197 train_loss: 0.2205 train_acc: 93.2374
Epoch: 198 train_loss: 0.2714 train_acc: 91.7986
Epoch: 199 train_loss: 0.2891 train_acc: 91.9424
Epoch: 200 train_loss: 0.2700 train_acc: 90.9352
Epoch: 201 train_loss: 0.2510 train_acc: 92.2302
Epoch: 202 train_loss: 0.2557 train_acc: 90.9352
Epoch: 203 train_loss: 0.2439 train_acc: 92.3741
Epoch: 204 train_loss: 0.2199 train_acc: 92.3741
Epoch: 205 train_loss: 0.2061 train_acc: 93.8130
Epoch: 206 train_loss: 0.2101 train_acc: 94.6763
Epoch: 207 train_loss: 0.2261 train_acc: 92.9496
Epoch: 208 train_loss: 0.2675 train_acc: 92.3741
Epoch: 209 train_loss: 0.2212 train_acc: 93.8130
Epoch: 210 train_loss: 0.2251 train_acc: 93.8130
Epoch: 211 train_loss: 0.2255 train_acc: 93.3813
Epoch: 212 train_loss: 0.1906 train_acc: 94.8201
Epoch: 213 train_loss: 0.1993 train_acc: 94.9640
Epoch: 214 train_loss: 0.2196 train_acc: 93.5252
Epoch: 215 train_loss: 0.1859 train_acc: 94.2446
Epoch: 216 train_loss: 0.1960 train_acc: 94.3885
Epoch: 217 train_loss: 0.1975 train_acc: 94.2446
Epoch: 218 train_loss: 0.2274 train_acc: 93.9568
Epoch: 219 train_loss: 0.2261 train_acc: 92.3741
Epoch: 220 train_loss: 0.2016 train_acc: 94.3885
Epoch: 221 train_loss: 0.1827 train_acc: 95.3957
Epoch: 222 train_loss: 0.2437 train_acc: 91.9424
Epoch: 223 train_loss: 0.2060 train_acc: 93.9568
Epoch: 224 train_loss: 0.1887 train_acc: 94.6763
Epoch: 225 train_loss: 0.1982 train_acc: 94.5324
Epoch: 226 train_loss: 0.1992 train_acc: 94.2446
Epoch: 227 train_loss: 0.1813 train_acc: 94.8201
Epoch: 228 train_loss: 0.1900 train_acc: 94.6763
Epoch: 229 train_loss: 0.1830 train_acc: 94.5324
Epoch: 230 train_loss: 0.1804 train_acc: 95.3957
Epoch: 231 train_loss: 0.1598 train_acc: 96.8345
Epoch: 232 train_loss: 0.1606 train_acc: 96.1151
Epoch: 233 train_loss: 0.1919 train_acc: 94.1007
Epoch: 234 train_loss: 0.1649 train_acc: 95.5396
Epoch: 235 train_loss: 0.1981 train_acc: 93.9568
Epoch: 236 train_loss: 0.1800 train_acc: 95.2518
Epoch: 237 train_loss: 0.1698 train_acc: 95.6834
Epoch: 238 train_loss: 0.1808 train_acc: 94.9640
Epoch: 239 train_loss: 0.1985 train_acc: 94.1007
Epoch: 240 train_loss: 0.2011 train_acc: 94.2446
Epoch: 241 train_loss: 0.1652 train_acc: 96.1151
Epoch: 242 train_loss: 0.1733 train_acc: 95.6834
Epoch: 243 train_loss: 0.1889 train_acc: 94.9640
Epoch: 244 train_loss: 0.1638 train_acc: 95.6834
Epoch: 245 train_loss: 0.1528 train_acc: 97.1223
Epoch: 246 train_loss: 0.1507 train_acc: 97.1223
Epoch: 247 train_loss: 0.1499 train_acc: 96.5468
Epoch: 248 train_loss: 0.1634 train_acc: 95.5396
Epoch: 249 train_loss: 0.1469 train_acc: 95.9712
Epoch: 250 train_loss: 0.1839 train_acc: 93.8130
Epoch: 251 train_loss: 0.2340 train_acc: 92.2302
Epoch: 252 train_loss: 0.2215 train_acc: 93.5252
Epoch: 253 train_loss: 0.1855 train_acc: 95.1079
Epoch: 254 train_loss: 0.1560 train_acc: 96.2590
Epoch: 255 train_loss: 0.1420 train_acc: 96.2590
Epoch: 256 train_loss: 0.1640 train_acc: 95.6834
Epoch: 257 train_loss: 0.1531 train_acc: 96.4029
Epoch: 258 train_loss: 0.1295 train_acc: 96.4029
Epoch: 259 train_loss: 0.1438 train_acc: 96.4029
Epoch: 260 train_loss: 0.1526 train_acc: 95.8273
Epoch: 261 train_loss: 0.1775 train_acc: 95.5396
Epoch: 262 train_loss: 0.1445 train_acc: 97.1223
Epoch: 263 train_loss: 0.1511 train_acc: 95.6834
Epoch: 264 train_loss: 0.1831 train_acc: 94.6763
Epoch: 265 train_loss: 0.1407 train_acc: 96.6906
Epoch: 266 train_loss: 0.1613 train_acc: 96.5468
Epoch: 267 train_loss: 0.1330 train_acc: 97.2662
Epoch: 268 train_loss: 0.1262 train_acc: 97.2662
Epoch: 269 train_loss: 0.1472 train_acc: 95.9712
Epoch: 270 train_loss: 0.1363 train_acc: 96.9784
Epoch: 271 train_loss: 0.1484 train_acc: 96.4029
Epoch: 272 train_loss: 0.1383 train_acc: 96.2590
Epoch: 273 train_loss: 0.1619 train_acc: 95.6834
Epoch: 274 train_loss: 0.1311 train_acc: 96.6906
Epoch: 275 train_loss: 0.1389 train_acc: 96.8345
Epoch: 276 train_loss: 0.1248 train_acc: 97.6978
Epoch: 277 train_loss: 0.1246 train_acc: 96.6906
Epoch: 278 train_loss: 0.1371 train_acc: 97.2662
Epoch: 279 train_loss: 0.1516 train_acc: 95.8273
Epoch: 280 train_loss: 0.1434 train_acc: 96.5468
Epoch: 281 train_loss: 0.1295 train_acc: 96.6906
Epoch: 282 train_loss: 0.1356 train_acc: 96.2590
Epoch: 283 train_loss: 0.1280 train_acc: 96.6906
Epoch: 284 train_loss: 0.1354 train_acc: 97.1223
Epoch: 285 train_loss: 0.1521 train_acc: 95.8273
Epoch: 286 train_loss: 0.1545 train_acc: 96.1151
Epoch: 287 train_loss: 0.1636 train_acc: 96.6906
Epoch: 288 train_loss: 0.1540 train_acc: 95.8273
Epoch: 289 train_loss: 0.1237 train_acc: 96.9784
Epoch: 290 train_loss: 0.1246 train_acc: 97.1223
Epoch: 291 train_loss: 0.1395 train_acc: 95.8273
Epoch: 292 train_loss: 0.1313 train_acc: 97.2662
Epoch: 293 train_loss: 0.1327 train_acc: 96.9784
Epoch: 294 train_loss: 0.1272 train_acc: 97.6978
Epoch: 295 train_loss: 0.1237 train_acc: 96.8345
Epoch: 296 train_loss: 0.1127 train_acc: 97.6978
Epoch: 297 train_loss: 0.1164 train_acc: 97.4101
Epoch: 298 train_loss: 0.1421 train_acc: 96.2590
Epoch: 299 train_loss: 0.1458 train_acc: 95.5396
Epoch: 300 train_loss: 0.1333 train_acc: 96.6906
Running Time: 227.97
------------------------------
Final result:
OA: 0.7518 | AA: 0.8448 | Kappa: 0.7196
[0.67919075 0.84311224 0.92391304 0.87248322 0.8651363 0.92027335
0.76579521 0.56823821 0.68439716 0.99382716 0.914791 0.76363636
0.97777778 0.74358974 1. 1. ]
------------------------------
Parameter:
dataset: Indian
flag_test: train
mode: CAF
gpu_id: 0
seed: 0
batch_size: 64
test_freq: 5
patches: 7
band_patches: 3
epoches: 300
learning_rate: 0.0005
gamma: 0.9
weight_decay: 0.005
—
Reply to this email directly, view it on GitHub
<#12 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AFL2GZTS42V2KAPO7ZXHDZDVKOFNNANCNFSM5WBJ3ROQ>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
好的，老师，谢谢您的回复 |
老师你好，我在训练时使用的是这个命令：python demo.py --dataset='Indian' --epoches=300 --patches=7 --band_patches=3 --mode='CAF' --weight_decay=5e-3，但是跑出来的精度只有75%，方便帮我看一下可能是哪里出现的问题吗？
The text was updated successfully, but these errors were encountered: